While myotonic dystrophy (DM) is classified as a muscular dystrophy, it is a progressive multi-systemic disease that has significant effects on central nervous system (CNS) function. These CNS manifestations include mental retardation and autistic behaviors in congenital DM and hypersomnia, executive dysfunction and cortical atrophy in the adult-onset form. Studies on the molecular basis of DM have revealed a novel disease mechanism, RNA-mediated pathogenesis, that involves the expansion of microsatellite C(C)TG repeats in two unrelated genes and the synthesis of toxic C(C)UG expansion RNAs. These pathogenic RNAs interfere with the normal alternative splicing functions of the CELF and MBNL proteins resulting in the persistence of, or reversion to, fetal isoforms in adult tissues. To determine if this RNA-mediated disease model is relevant to CNS dysfunction in DM, we have recently generated Mbnl2 knockout mice which recapitulate characteristic pathological features of the DM brain including abnormal REM sleep propensity and spatial memory deficits. In this proposal, we will first test the MBNL2 sequestration hypothesis for DM relevant CNS pathogenesis by demonstrating that transfer of MBNL2 from its normal RNA targets to C(C)UG repeats occurs in the DM brain. Next, we will determine if Mbnl2 knockout mice model the full range of DM CNS features and link specific mis-splicing events to characteristic disease manifestations. Finally, we will test the hypothesis that interactions between MBNL genes are disrupted in the congenital disease. These studies will provide novel animal models of adult-onset and congenital DM for future disease mechanism studies and the development of therapies that target the DM brain.